Machine Learning Discoveries of DNA Repair-X Synergy in ETC-1922159 Treated Colorectal Cancer Cells

1. Introduction

In the unpublished preprint Sinha [1], a frame work of a search engine was developed which can rank combinations of factors (genes/proteins) in a signaling pathway. Such combinations are of import due to the vast search space in which they exist and the difficulty to find them. The search engine facilitates in prioritizing the combinations as ranked biological hypotheses which the biologists might want to test in wet lab, to know if a synergistic combination is prevalent in a signaling pathway, in a direct or indirect manner. Interested readers are advised to go through unpublished preprints Sinha [1] and Sinha [2] for details regarding the search engine and the discoveries mentioned in there.

2. Materials and Methods

2.1. Combinatorial Search Problem and a Possible Solution

The issue of combinatorial search problem and a possible solution has been addressed in Sinha [3] and Sinha [2]. The details of the methodology of this manuscript have been explained in great detail in Sinha [3] & its application in Sinha [2]. Readers are requested to go through the same for gaining deeper insight into the working of the pipeline and its use of published data set generated after administration of ETC-1922159. In order to understand the significance of the solution proposed to the problem of combinatorial search that the biologists face in revealing unknown biological search problem, these works are of importance.

Briefly, from Sinha [2], the pipleline works by computing sensitivity indicies for each of these unique combinations and then vectorising these indices to connote and form discriminative feature vector for each combination. Since each combination is unique, the training and the test data are same. In the training data, the combinations are arranged and ranks from 1 to n are assigned. The ranking algorithm then learns the patterns from these combinations/sensitivity index vectors. Next the learned model is used to rank the test data by generating the ranking score for each of the unique combination. Sorting these shuffled scores of test data leads to prioritization of the combinations. Joachims [4] show an example of applying learned model to training data (same as the test data) in https://www.cs.cornell.edu/people/tj/svm_light/svm_rank.html. Note that these combinations are now ranked and give the biologists a chance to narrow down their focus on crucial biological hypotheses in the form of combinations which the biologists might want to test. Analogous to the webpage search engine, where the click of a button for a few key-words leads to a ranked list of web links, the pipeline uses sensitivity indices as an indicator of the strength of the influence of factors or their combinations, as a criteria to rank the combinations.

3. Results & Discussion

3.1. DNA Repair Related Synergies

3.1.1. XRCC - RAD Cross Family Analysis

X-ray repair cross-complementing protein (XRCC) plays major role in DNA repair process, especially in Double Strand Repair (DBS) Thacker and Zdzienicka [5] and Thacker and Zdzienicka [6]. Sultana et al. [7] observe that ataxia telangiectasia mutated and RAD3 related (ATR) protein kinase inhibition is synthetically lethal in XRCC1 deficient ovarian cancer cells. Della-Maria et al. [8] observe that human Mre11/human RAD50/Nbs1 and DNA ligase III$\alpha$/XRCC1 protein complexes act together in an alternative nonhomologous end joining pathway. These findings along with multiple published work indicate the joint synergy of XRCC - RAD family. In colorectal cancer cell lines treated with ETC-1922159, both XRCC and RAD members were found to be down regulated. The search engine gave the 2^nd order synergies between XRCC - RAD families, low numerical valued ranks to signify plausible synergistic down regulations that might not have been explored. Table 1 shows the rankings of RAD family w.r.t XRCC family and Table 2 shows the rankings of the XRCC family w.r.t RAD family. In Table 1 we found RAD-18/51/51AP1/51C/54B/54L to be down regulated w.r.t XRCC1. These are reflected with rankings of 1027 (laplace), 456 (linear) and 1355 (rbf) for RAD-18 - XRCC1; 282 (laplace), 365 (linear) and 1003 (rbf) for RAD51 - XRCC1; 753 (laplace), 5 (linear) and 27 (rbf) for RAD51AP1 - XRCC1; 337 (laplace), 111 (linear) and 968 (rbf) for RAD51C - XRCC1; 175 (laplace), 224 (linear) and 78 (rbf) for RAD54B - XRCC1; and 327 (laplace), 889 (linear) and 709 (rbf) for RAD54L - XRCC1. RAD-18/51/51AP1/51C/54B/54L were also found to be down regulated w.r.t XRCC2. These are reflected in 1388 (laplace), 847 (linear) and 765 (rbf) for XRCC2 - RAD18; 1247 (laplace), 1033 (linear) and 629 (rbf) for XRCC2 - RAD51; 302 (laplace); 247 (linear) and 42 (rbf) for XRCC2 - RAD51AP1; 1079 (laplace), 674 (linear) and 323 (rbf) for XRCC2 - RAD51C; 387 (laplace), 566 (linear) and 506 (rbf) for XRCC2 - RAD54B; and 976 (laplace), 918 (linear) and 847 (rbf) for XRCC2 - RAD54L. RAD-18/51/51AP1/51C/54B/54L were found to be down regulated with w.r.t XRCC6. These are reflected in 541 (laplace), 25 (linear) and 1068 (rbf) for RAD18 - XRCC6; 608 (laplace), 425 (linear) and 900 (rbf) for RAD51 - XRCC6; 216 (laplace), 67 (linear) and 83 (rbf) for RAD51AP1 - XRCC6; 426 (laplace), 865 (linear) and 503 (rbf) for RAD51C - XRCC6; 3 (laplace), 610 (linear) and 112 (rbf) for RAD54B - XRCC6; and 85 (laplace), 252 (linear) and 432 (rbf) for RAD54L - XRCC6. RAD-1/18/50/51/51AP1/51C/54B/54L were found to be down regulated w.r.t XRCC6BP1. These are reflected in 1167 (laplace) and 308 (rbf) for RAD1 - XRCC6BP1; 656 (linear) and 1612 (rbf) for RAD18 - XRCC6BP1; 1302 (laplace) and 328 (rbf) for XRCC6BP1 - RAD50; 435 (laplace), 495 (linear) and 1275 (rbf) for RAD51 - XRCC6BP1; 81 (laplace), 177 (linear) and 73 (rbf) for RAD51AP1 - XRCC6BP1; 645 (laplace), 1366 (linear) and 1414 (rbf) for RAD51C - XRCC6BP1; 154 (laplace), 693 (linear) and 1398 (rbf) for RAD54B - XRCC6BP1; and 420 (linear) and 1060 (rbf) for RAD54L - XRCC6BP1;

In Table 2 we found XRCC-2/6BP1 to be down regulated w.r.t RAD1. These are reflected in 62 (laplace), 498 (linear) and 1231 (rbf) for RAD1 - XRCC2; and 764 (laplace) and 1325 (rbf) for RAD1 - XRCC6BP1. XRCC-1/2/6 were found to be down regulated with w.r.t RAD18. These are reflected in 927 (laplace) and 200 (rbf) for RAD18 - XRCC1; 506 (laplace) and 1517 (rbf) for RAD18 - XRCC2; and 279 (laplace) and 804 (rbf) for RAD18 - XRCC6; XRCC-2/6BP1 were found to be down regulated w.r.t RAD50. These are reflected in rankings of 53 (laplace), 244 (linear) and 147 (rbf) for XRCC-2 - RAD50; and 1375 (linear) and 1366 (rbf) for RAD50 - XRCC6BP1. XRCC-6/6BP1 were found to be down regulated w.r.t RAD51; These are reflected in rankings of 80 (laplace) and 1244 (linear) for XRCC6 - RAD51; and 792 (laplace), 951 (linear) and 1595 (rbf) for XRCC6BP1 - RAD51. XRCC-2/6BP1 were found to be down regulated w.r.t RAD51AP1. These were reflected in 78 (laplace), 112 (linear) and 351 (rbf) for XRCC2 - RAD51AP1; and 936 (linear) and 974 (rbf) for XRCC6BP1 - RAD51AP1; XRCC2 was found to be down regulated w.r.t RAD51C. This are reflected in 1695 (laplace), 932 (linear) and 520 (rbf) for XRCC2 - RAD51C. XRCC2 was found to be down regulated w.r.t RAD54B. This is reflected in rankings of 1554 (laplace), 744 (linear) and 620 (rbf) for XRCC2 - RAD54B. XRCC-1/2/6/6BP1 were found to be down regulated w.r.t RAD54L. These are reflected in rankings of 657 (linear) and 525 (rbf) for XRCC1 - RAD54L; 167 (laplace) and 565 (rbf) for XRCC2 - RAD54L; 496 (linear) and 1247 (rbf) for XRCC6 - RAD54L; and 1389 (laplace), 1227 (linear) and 1454 (rbf) for RAD54L - XRCC6BP1;

Table 3 shows the derived influences which can be represented graphically, with the following influences - • RAD w.r.t XRCC with RAD-18/51/51AP1/51C/54B/54L $<-$ XRCC1; RAD-18/51/51AP1/51C/54B/54L $<-$ XRCC2; RAD-18/51/51AP1/51C/54B/54L $<-$ XRCC6 and RAD-1/18/50/51/51AP1/51C/54B/54L $<-$ XRCC6BP1; •; XRCC w.r.t RAD with RAD1 $->$ XRCC-2/6BP1; RAD18 $->$ XRCC-1/2/6; RAD50 $->$ XRCC-2/6BP1; RAD51 $->$ XRCC-6/6BP1; RAD51AP1 $->$ XRCC-2/6BP1; RAD51C $->$ XRCC-2; RAD54B $->$ XRCC-2; RAD54L $->$ XRCC-1/2/6/6BP1;

3.1.2. XRN2 - RAD Cross Family Analysis

XRN2 (5’-3’ exoribonuclease 2) is involved in RNA synthesis/trafficking and termination. Morales et al. [9] observe that XRN2 links transcription termination to DNA damage and replication stress. They found an increase in the amount of RAD51 foci in shXRN2 cells compared to controls, suggesting that cells depleted of XRN2 are subjected to an increased level of basal DNA damage and show that loss of XRN2 also leads to the focal accumulation of several factors required for homologous recombination, such as ATM, BRCA1 and RAD51. This definitely shows that there is synergy between the XRN2 and RAD51. We found that both the XRN2 and RAD families were down regulated in CRC cell after ETC-1922159 treatment. The search engine gave rankings to the combinations of the XRN2 and RAD family members with low numerical valued in silico ranks. Table 4 shows the rankings of XRN2 w.r.t RAD family and vice versa. Following this is the derived influences in Table 5. We find RAD-51AP1/51/54L/51C/18/54B to be down regulated w.r.t XRN2. These are reflected in rankings of 340 (laplace), 545 (linear) and 290 (rbf) for RAD51AP1 - XRN2; 387 (laplace), 560 (linear) and 605 (rbf) for XRN2 - RAD51; 594 (laplace), 827 (linear) and 879 (rbf) for XRN2 - RAD54L; 639 (laplace), 1236 (linear) and 745 (rbf) for XRN2 - RAD51C; 794 (laplace), 688 (linear) and 804 (rbf) for XRN2 - RAD18; 255 (linear) and 122 (rbf) for XRN2 - RAD1 and 951 (laplace), 165 (linear) and 34 (rbf) for XRN2 - RAD54B; On the other hand, XRN2 was found to be down regulated w.r.t RAD family. These are reflected in rankings of 255 (laplace) and 122 (rbf) for XRN2 - RAD1; 1256 (linear) and 852 (rbf) for XRN2 - RAD51AP1; 1541 (laplace) and 1246 (linear) for XRN2 - RAD54L and 1037 (laplace) and 1777 (linear) for XRN2 - RAD51C. Graphical depiction of XRN2 and RAD family dependencies is shown as • RAD w.r.t XRN2 with XRN2 $->$ RAD-51AP1/51/54L/51C/18/54B and • XRN2 w.r.t RAD with XRN2 $<-$ RAD1; XRN2 $<-$ RAD51AP1; XRN2 $<-$ RAD54L; XRN2 $<-$ RAD51C;

Table 5 shows the derived influences which can be represented graphically, with the following influences - • RAD w.r.t XRN2 with XRN2 $->$ RAD-51AP1/51/54L/51C/18/54B; and • XRN2 w.r.t RAD with XRN2 $<-$ RAD-1/51AP1/54L/51C.

3.1.3. NKRF - RAD Cross Family Analysis

Not much is known about the NKRF (NF-$\kappa$B-repressing factor) and RAD members. We found the combinations to be down regulated by the search engine between NKRF and RAD family. Table 6 shows the rankings of NKRF and RAD family. We found NKRF down regulated w.r.t RAD family. These are reflected in rankings of 1724 (laplace), 1642 (linear) and 649 (rbf) for RAD51AP1 $<-$ NKRF; 982 (laplace), 1724 (linear) and 1352 (rbf) RAD51 $<-$ NKRF; 1727 (laplace), 1387 (linear) and 1120 (rbf) for RAD54L $<-$ NKRF; 1568 (laplace), 472 (linear) and 1505 (rbf) for RAD51C $<-$ NKRF; 1508 (laplace), 615 (linear) and 405 (rbf) for RAD18 $<-$ NKRF; and 1476 (laplace), 1189 (linear) and 1534 (rbf) for RAD54B $<-$ NKRF;

Also, we found RAD family to be down regulated w.r.t NKRF. These are reflected in rankings of 157 (laplace) and 553 (linear) for RAD51AP1 - NKRF; 439 (laplace), 1441 (linear) and 1606 (rbf) for RAD51 - NKRF; 117 (laplace), 1175 (linear) and 1415 (rbf) for RAD54L - NKRF; 418 (laplace), and 1653 (rbf) for RAD51C - NKRF; 164 (laplace) and 1509 (rbf) for RAD18 - NKRF; 1391 (laplace), 1115 (linear) and 735 (rbf) NKRF - RAD1; 1354 (laplace), 851 (linear) and 824 (rbf) for NKRF - RAD50;

Table 7 shows the derived influences which can be represented graphically, with the following influences - • RAD w.r.t NKRF with RAD51AP1 $<-$ NKRF; RAD51 $<-$ NKRF; RAD54L $<-$ NKRF; RAD51C $<-$ NKRF; RAD18 $<-$ NKRF; RAD1 $<-$ NKRF; RAD54B $<-$ NKRF and • NKRF w.r.t RAD with RAD51AP1 $->$ NKRF; RAD51 $->$ NKRF; RAD54L $->$ NKRF; RAD51C $->$ NKRF; RAD18 $->$ NKRF; NKRF $->$ RAD1; NKRF $->$ RAD50.

3.1.4. RAD - BCL Cross Family Analysis

Saintigny et al. [10] show a specific role of BCL2 in suppression of the RAD51 recombination pathway. They observe that BCL2 consistently inhibits recombination stimulated by RAD51 overexpression and alters RAD51 protein by post-translation modification. Based on the findings that CARD9 and BCL10 acted together to activate NF-kB following cytosolic DNA sensing, Meng et al. [11] demonstrated that BCL10 was recruited to the dsDNA–RAD50 complexes in a CARD9-dependent manner. These mechanisms point to a synergy between BCL and RAD family. In CRC cells treated with ETC-1922159, BCL and RAD family members were found to be down regulated. The search engine alloted the combinations of RAD and BCL low numerical valued ranks pointing to possible synergistic down regulations. Table 8 shows rankings of BCL and RAD w.r.t to each other. The left half of the table points to rankings of BCL family w.r.t RAD family. The right half of the table points to rankings of RAD family w.r.t BCL family.

On the left side, BCL2L12 was found to be down regulated w.r.t RAD-1/18/50/51/51C/54B/54L. These are reflected in rankings of 1530 (linear) and 1401 (rbf) for RAD1 - BCL2L12; 675 (laplace) and 1312 (rbf) for RAD18 - BCL2L12; 1151 (linear) and 929 (rbf) for RAD50 - BCL2L12; 1234 (laplace) and 1334 (linear) for RAD51 - BCL2L12; 1561 (laplace) and 1647 (rbf) for RAD51C - BCL2L12; 1329 (linear) and 1625 (rbf) for RAD54B - BCL2L12, and 821 (linear) and 210 (rbf) for RAD54L - BCL2L12; BCL6B was found to be down regulated w.r.t RAD-1/18/50/51/51AP1/51C/54B/54L. 194 (laplace), 481 (linear) and 102 (rbf) for RAD1 - BCL6B; 176 (linear) and 929 (rbf) for RAD18 - BCL6B; 860 (laplace), 87 (linear) and 74 (rbf) for RAD50 - BCL6B; 263 (linear) and 58 (rbf) for RAD51 - BCL6B; 723 (laplace), 428 (linear) and 579 (rbf) for RAD51AP1 - BCL6B; 660 (laplace), 521 (linear) and 1609 (rbf) for RAD51C - BCL6B; 708 (laplace), 596 (linear) and 647 (rbf) for RAD54B - BCL6B; and 108 (laplace) and 1326 (rbf) for RAD54L - BCL6B; BCL7A was found to be down regulated w.r.t RAD-1/18/50/51/54L. These are reflected in rankings of 690 (laplace) and 1202 (rbf) for BCL7A - RAD1; 385 (laplace) and 185 (rbf) for BCL7A - RAD18; 137 (laplace), 601 (linear) and 41 (rbf) for RAD50 - BCL7A; 514 (laplace) and 1694 (linear) for BCL7A - RAD51; 1519 (laplace), 418 (linear) and 842 (rbf) for RAD54L - BCL7A; BCL9 was found to be down regulated w.r.t RAD-18/51/51C/54L. These are reflected in rankings for 461 (laplace) and 1453 (linear) for RAD18 - BCL9; 1143 (linear) and 95 (rbf) for RAD51 - BCL9; 956 (laplace) and 376 (rbf) for RAD51C - BCL9; 1450 (laplace), 1096 (linear) and 400 (rbf) for RAD54L - BCL9; BCL11A was found to be down regulated w.r.t RAD-1/18/50/51/51AP1/51C/54B. These are reflected in rankings of 1069 (laplace), 507 (linear) and 1267 (rbf) for RAD1 - BCL11A; 1561 (laplace), 169 (linear) and 692 (rbf) for RAD18 - BCL11A; 582 (laplace), 1144 (linear) and 1047 (rbf) for RAD50 - BCL11A; 1120 (laplace), 752 (linear) and 645 (rbf) for RAD51AP1 - BCL11A; 1024 (laplace), 199 (linear) and 899 (rbf) for RAD51C - BCL11A; and 1037 (laplace), 917 (linear) and 867 (rbf) for RAD54B - BCL11A. BCL11B was found to be down regulated w.r.t RAD-50/51/51AP1/54B/54L. These are reflected in rankings of 1198 (linear) and 903 (rbf) for RAD50 - BCL11B; 449 (linear) and 971 (rbf) for RAD51 - BCL11B; 1247 (laplace), 908 (linear) and 1671 (rbf) for RAD51AP1 - BCL11B; 1193 (laplace), 1192 (linear) and 832 (rbf) for RAD54B - BCL11B and 1421 (laplace) and 1385 (linear) for RAD54L - BCL11B.

On the right side, w.r.t BCL2L12, RAD-18/50/51/51AP1/51C/54B/54L were found to be down regulated. These are found in the rankings of 779 (laplace), 652 (linear) and 1388 (rbf) for RAD18 - BCL2L12; 1668 (laplace), 2566 (linear) and 1703 (rbf) for RAD50 - BCL2L12; 1164 (laplace), 365 (linear), 1213 (rbf) for RAD51 - BCL2L12; 306 (laplace), 57 (linear) and 28 (rbf) for RAD51AP1 - BCL2L12; 495 (laplace), 1191 (linear) and 429 (rbf) for RAD51C - BCL2L12; 678 (laplace), 432 (linear) and 787 (rbf) for RAD54B - BCL2L12; and 901 (laplace), 1128 (linear) and 263 (rbf) for RAD54L - BCL2L12; w.r.t BCL6B, RAD-18/51/51AP1/51C/54B/54L were found to be down regulated. These are reflected in rankings of 1113 (laplace), 640 (linear) and 482 (rbf) for RAD18 - BCL6B; 287 (laplace), 681 (linear) and 497 (rbf) for RAD51 - BCL6B; 1607 (laplace), 1638 (linear) and 916 (rbf) for RAD51AP1 - BCL6B; 43 (laplace), 871 (linear) and 999 (rbf) for RAD51C - BCL6B; 1212 (laplace), 1392 (linear) and 1170 (rbf) for RAD54B - BCL6B; and 1009 (linear) and 785 (rbf) for RAD54L - BCL6B; w.r.t BCL7A, RAD-18/51/51AP1/51C/54B/54L were found to be down regulated. These are reflected in rankings of 1514 (laplace), 1515 (linear), 783 (rbf) for RAD18 - BCL7A; 879 (laplace), 274 (linear) and 639 (rbf) for RAD51 - BCL7A; 412 (laplace), 416 (linear) and 4 (rbf) for RAD51AP1 - BCL7A; 215 (laplace), 394 (linear) and 461 (rbf) for RAD51C - BCL7A; 809 (laplace), 1407 (linear) and 213 (rbf) for RAD54B - BCL7A and 435 (laplace), 783 (linear) and 1499 (rbf) for RAD54L - BCL7A. w.r.t BCL9, RAD-18/50/51/51AP1/51C/54B/54L were found to be down regulated. These are reflected in the rankings of 656 (laplace), 1194 (linear) and 482 (rbf) for RAD18 - BCL9; 1441 (linear) and 1098 (rbf) for RAD50 - BCL9; 622 (laplace), 929 (linear), 860 (rbf) for RAD51 - BCL9; 331 (laplace), 61 (linear) and 102 (rbf) for RAD51AP1 - BCL9; 1113 (laplace), 417 (linear) and 1154 (rbf) for RAD51C - BCL9; 1045 (laplace), 53 (linear) and 650 (rbf) for RAD54B - BCL9 and 636 (laplace), 602 (linear) and 934 (rbf) for RAD54L - BCL9. w.r.t BCL11A, RAD-1/18/50/51/51AP1/51C/54B/54L were found to be down regulated. These are reflected in 1430 (laplace), 1475 (linear) and 1584 (rbf) for RAD1 - BCL11A; 465 (laplace) and 164 (linear) for RAD18 - BCL11A; 875 (linear) and 1226 (rbf) for RAD50 - BCL11A; 659 (laplace), 388 (linear) and 496 (rbf) for RAD51AP1 - BCL11A; 363 (laplace), 1673 (linear) and 97 (rbf) for RAD51C - BCL11A; 581 (laplace) and 799 (rbf) for RAD54B - BCL11A; and 846 (laplace) and 209 (rbf) for RAD54L - BCL11A; w.r.t BCL11B, RAD-1/50/51/51AP1/51C/54B/54L were found to be down regulated. These are reflected in rankings of 230 (linear) and 1373 (rbf) RAD1 - BCL11B; 919 (laplace) and 860 (linear) for RAD50 - BCL11B; 1095 (laplace) and 1238 (linear) RAD51 - BCL11B; 196 (laplace) and 987 (rbf) for RAD51AP1 - BCL11B; 1122 (laplace) and 1161 (rbf) for RAD51C - BCL11Bl; 363 (laplace) and 1561 (rbf) for RAD54B - BCL11B; 579 (laplace), 2543 (linear) and 159 (rbf) for RAD54L - BCL11B.

Table 9 shows the derived influences which can be represented graphically, with the following influences - • RAD w.r.t BCL with RAD-18/50/51/51AP1/51C/54B/54L $<-$ BCL-2L12; RAD-18/51/51AP1/51C/54B/54L $<-$ BCL-6B; RAD-18/51/51AP1/51C/54B/54L $<-$ BCL-7A; RAD-18/50/51/51AP1/51C/54B/54L $<-$ BCL-9; RAD-1/18/50/51/51AP1/51C/54B/54L $<-$ BCL-11A; RAD-1/50/51/51AP1/51C/54B/54L $<-$ BCL-11B; and • BCL w.r.t RAD with RAD-1/18/50/51/51C/54B/54L $->$ BCL-2L12; RAD-1/18/50/51/51AP1/51C/54B/54L $->$ BCL-6B; RAD-1/18/50/51/54L $->$ BCL-7A; RAD-18/51/51C/54L $->$ BCL-9; RAD-1/18/50/51/51AP1/51C/54B $->$ BCL-11A; and RAD-50/51/51AP1/54B/54L $->$ BCL-11B.

3.1.5. RAD - EXOSC Cross Family Analysis

Marin-Vicente et al. [12] show that RRP6/EXOSC10 is required for the repair of DNA double-strand breaks by homologous recombination. The authors results suggest that ribonucleolytic activity of RRP6/EXOSC10 is required for the recruitment of RAD51 to DSBs. The therapeutic potential of exosome-mediated siRNA delivery was demonstrated in vitro by the strong knockdown of RAD51, a prospective therapeutic target for cancer cells (Shtam et al. [13]). These findings point to the synergy between EXOSC and RAD family. In CRC cells treated with ETC-1922159, they were down regulated and the search engine allocated low numerical rankings for combinations, thus pointing to possible synergistic down regulation. Table 10 shows the rankings of the EXOSC and RAD family w.r.t to each other. On the left half of the table is the rankings of EXOSC w.r.t RAD family. EXOSC2 was found to be down regulated w.r.t RAD-1/18/50/51/51AP1/51C/54B/54L. These are reflected in rankings of 1033 (laplace), 1311 (linear) and 1207 (rbf) for EXOSC2 - RAD1; 1210 (laplace) and 995 (linear) for EXOSC2 - RAD18; 1124 (laplace), 698 (linear) and 629 (rbf) for EXOSC2 - RAD50; 1754 (laplace), 191 (linear)and 633 (rbf) and for EXOSC2 - RAD51; 198 (laplace) and 1462 (linear) for EXOSC2 - RAD51AP1; 87 (laplace), 463 (linear) and 1130 (rbf) for EXOSC2 - RAD51C; 351 (laplace), 135 (linear) and 142 (rbf) for EXOSC2 - RAD54B; and 1131 (laplace), 1652 (linear) and 320 (rbf) for EXOSC2 - RAD54L. EXOSC3 was found to be down regulated w.r.t RAD-1/18/51/51AP1/54L. These are reflected in rankings of 1677 (linear) and 549 (rbf) for EXOSC3 - RAD1; 1676 (laplace) and 184 (rbf) for EXOSC3 - RAD18; 894 (laplace) and 1066 (linear) for EXOSC3 - RAD51; 1037 (linear) and 804 (rbf) for EXOSC3 - RAD51AP1, and 469 (linear) and 736 (rbf) for EXOSC3 - RAD54L. EXOSC5 was found to be down regulated w.r.t RAD-1/18/50/51/51AP1/51C/54B/54L. These are reflected in rankings of 568 (laplace), 1169 (linear) and 1699 (rbf) for EXOSC5 - RAD1; 219 (linear) and 1652 (rbf) for EXOSC5 - RAD18; 447 (laplace), 195 (linear) and 475 (rbf) for EXOSC5 - RAD50; 431 (linear) and 1121 (rbf) for EXOSC5 - RAD51; 1290 (laplace), 487 (linear) and 430 (rbf) for EXOSC5 - RAD51AP1; 1284 (laplace) and 1264 (linear) for EXOSC5 - RAD51C; 940 (laplace), 812 (linear) and 1036 (rbf) for EXOSC5 - RAD54B; and 408 (laplace) and 1407 (rbf) for EXOSC5 - RAD54L; EXOSC6 was found to be down regulated w.r.t RAD-18/51/54L. These were reflected in rankings of 1637 (laplace), 1599 (linear) and 2254 (rbf) for EXOSC6 - RAD18; 1056 (laplace), 1482 (linear) and 1007 (rbf) for EXOSC6 - RAD51; and 987 (laplace) and 1642 (rbf) for EXOSC6 - RAD54L; EXOSC7 was found to be down regulated w.r.t RAD-1/18/51C/54B/54L. These are reflected in rankings of 1735 (linear) and 1210 (rbf) for EXOSC7 - RAD1; 490 (laplace), 1688 (linear) and 1331 (rbf) for EXOSC7 - RAD18; 1113 (laplace), 1623 (linear) and 530 (rbf) for EXOSC7 - RAD51C; 1612 (linear) and 1191 (rbf) for EXOSC7 - RAD54B; and 1550 (laplace), 1754 (linear) and 1728 (rbf) for EXOSC7 - RAD54L; EXOSC8 was found to be down regulated w.r.t RAD-18/51/51AP1/54B/54L. These are reflected in 805 (laplace) and 1564 (rbf) for EXOSC8 - RAD18; 404 (laplace) and 1630 (linear) for EXOSC8 - RAD51; 1567 (linear) and 1701 (rbf) for EXOSC8 - RAD51AP1; 1562 (laplace) and 1736 (rbf) for EXOSC8 - RAD54B; and 1248 (laplace), 622 (linear) and 239 (rbf) for EXOSC8 - RAD54L; EXOSC9 was found to be down regulated w.r.t RAD-1/18/50/51/51C/54B/54L. These are reflected in rankings of 175 (linear) and 1648 (rbf) for EXOSC9 - RAD1; 1533 (laplace), 774 (linear) and 1180 (rbf) for EXOSC9 - RAD18; 545 (laplace), 183 (linear) and 467 (rbf) for EXOSC9 - RAD50; 866 (laplace), 106 (linear) and 99 (rbf) for EXOSC9 - RAD51; 110 (laplace), 742 (linear) and 200 (rbf) for EXOSC9 - RAD51C; 179 (laplace), 178 (linear) and 84 (rbf) for EXOSC9 - RAD54B and 1113 (laplace) and 22 (rbf) for EXOSC9 - RAD54L;

On the right half of the table is the rankings of RAD family w.r.t EXOSC. RAD-18/51/51C/54B/54L was found to be down regulated w.r.t EXOSC2. These are reflected in rankings of 1115 (laplace), 979 (linear) and 654(rbf) for EXOSC2 - RAD18; 795 (laplace), 1332 (linear) and 441(rbf) for EXOSC2 - RAD51; 636 (laplace), 564 (linear) and 152(rbf) for EXOSC2 - RAD51C; 278 (laplace), 132 (linear) and 282(rbf) for EXOSC2 - RAD54B and 125 (laplace), 888 (linear) and 545(rbf) for EXOSC2 - RAD54L. RAD-18/50/51/51AP1/51C/54B/54L was found to be down regulated w.r.t EXOSC3. These are reflected in rankings of 1468 (linear) and 767 (rbf) for EXOSC3 - RAD18; 1062 (laplace) and 596 (linear) for EXOSC3 - RAD50; 727 (laplace), 583 (linear) and 963 (rbf) for EXOSC3 - RAD51; 100 (laplace), 49 (linear) and 219 (rbf) for EXOSC3 - RAD51AP1; 663 (laplace), 869 (linear) and 887 (rbf) for EXOSC3 - RAD51C; 384 (laplace), 277 (linear) and 310 (rbf) for EXOSC3 - RAD54B and 546 (laplace), 1117 (linear) and 808 (rbf) for EXOSC3 - RAD54L; RAD-1/18/51/51AP1/51C/54B/54L was found to be down regulated w.r.t EXOSC5. These are reflected in rankings of 1716 (linear) and 1718 (rbf) for EXOSC5 - RAD1; 1026 (laplace), 550 (linear) and 253 (rbf) for EXOSC5 - RAD18; 260 (laplace), 1095 (linear) and 137 (rbf) for EXOSC5 - RAD51; 1555 (laplace) and 976 (rbf) for EXOSC5 - RAD51AP1; 233 (laplace), 1003 (linear) and 359 (rbf) for EXOSC5 - RAD51C; 834 (laplace), 1825 (linear) and 335 (rbf) for EXOSC5 - RAD54B; and 248 (laplace), 197 (linear) and 39 (rbf) for EXOSC5 - RAD54L. RAD-1/18/50/51AP1/51C/54L was found to be down regulated w.r.t EXOSC6. These are reflected in rankings of 142 (linear) and 639(rbf) for EXOSC6 - RAD1; 1118 (laplace), 1313 (linear) and 1549(rbf) for EXOSC6 - RAD18; 1722 (linear) and 575(rbf) for EXOSC6 - RAD50; 149 (laplace) and 1060 (linear) for EXOSC6 - RAD51AP1; 500 (laplace) and 1628 (linear) for EXOSC6 - RAD51C; and 885 (laplace), 271 (linear) and 1224(rbf) for EXOSC6 - RAD54L; RAD-18/51/51AP1/51C/54B/54L was found to be down regulated w.r.t EXOSC7. These were reflected in rankings of 441 (laplace), 385 (linear) and 1542(rbf) for EXOSC7 - RAD18; 376 (laplace), 1180 (linear) and 550(rbf) for EXOSC7 - RAD51; 35 (laplace), 97 (linear) and 786(rbf) for EXOSC7 - RAD51AP1; 854 (laplace), 671 (linear) and 1459(rbf) for EXOSC7 - RAD51C; 458 (laplace), 260 (linear) and 646(rbf) for EXOSC7 - RAD54B; and 464 (laplace), 528 (linear) and 790(rbf) for EXOSC7 - RAD54L; RAD-1/18/51/51AP1/51C/54B/54L was found to be down regulated w.r.t EXOSC8. These were reflected in rankings of 151 (linear) and 1563 (rbf) for EXOSC8 - RAD1; 764 (laplace), 523 (linear) and 29 (rbf) for EXOSC8 - RAD18; 98 (laplace), 1161 (linear) and 902 (rbf) for EXOSC8 - RAD51; 408 (laplace) and 541 (rbf) for EXOSC8 - RAD51AP1; 906 (laplace), 738 (linear) and 1052 (rbf) for EXOSC8 - RAD51C; 23 (laplace), 1578 (linear) and 130 (rbf) for EXOSC8 - RAD54B; and 651 (laplace), 1384 (linear) and 1047 (rbf) for EXOSC8 - RAD54L; RAD-1/18/50/51/51AP1/51C/54B/54L was found to be down regulated w.r.t EXOSC9. These were reflected in rankings of 1335 (laplace) and 978 (rbf) for EXOSC9 - RAD1; 54 (linear) and 540 (rbf) for EXOSC9 - RAD18; 211 (laplace) and 1377 (rbf) for EXOSC9 - RAD50; 807 (laplace), 74 (linear) and 429 (rbf) for EXOSC9 - RAD51; 103 (linear), 1210 (rbf) for EXOSC9 - RAD51AP1; 399 (laplace), 844 (linear) and 69 (rbf) for EXOSC9 - RAD51C; 466 (linear), 1286 (rbf) for EXOSC9 - RAD54B; and 536 (laplace), 724 (linear) and 414 (rbf) for EXOSC9 - RAD54L;

Table 11 shows the derived influences which can be represented graphically, with the following influences - • RAD w.r.t EXOSC with EXOSC-2 $->$ RAD-18/51/51C/54B/54L; EXOSC-3 $->$ RAD-18/50/51/51AP1/51C/54B/54L; EXOSC-5 $->$ RAD-1/18/51/51AP1/51C/54B/54L; EXOSC-6 $->$ RAD-1/18/50/51AP1/51C/54L; EXOSC-7 $->$ RAD-18/51/51AP1/51C/54B/54L; EXOSC-8 $->$ RAD-1/18/51/51AP1/51C/54B/54L; EXOSC-9 $->$ RAD-1/18/50/51/51AP1/51C/54B/54L; and • EXOSC w.r.t RAD with EXOSC-2 $<-$ RAD-1/18/50/51/51AP1/51C/54B/54L; EXOSC-3 $<-$ RAD-1/18/51/51AP1/54L; EXOSC-5 $<-$ RAD-1/18/50/51/51AP1/51C/54B/54L; EXOSC-6 $<-$ RAD-18/51/54L; EXOSC-7 $<-$ RAD-1/18/51C/54B/54L; EXOSC-8 $<-$ RAD-18/51/51AP1/54B/54L; and EXOSC-9 $<-$ RAD-1/18/50/51/51C/54B/54L.

3.1.6. XRCC - EXOSC cross family analysis

Not much is known about XRCC - EXOSC synergy, however both were found to be down regulated in CRC cells after treatment with ETC-1922159. The search engine also allocated rankings of low numerical values to several combinations thus indicating plausible synergistic down regulations. Table 12 shows the rankings of XRCC vs EXOSC family members.

On the left half of the table is the rankings of EXOSC w.r.t XRCC family. EXOSC2 was found to be down regulated w.r.t XRCC-1/2/6/6BP1. These are reflected in rankings of 277 (laplace), 176 (linear) and 423 (rbf) for EXOSC2 - XRCC1; 8 (laplace), 38 (linear) and 100 (rbf) for EXOSC2 - XRCC2; 1252 (laplace), 398 (linear) and 623 (rbf) for EXOSC2 - XRCC6; and 935 (laplace) and 905 (linear) for EXOSC2 - XRCC6BP1; EXOSC3 was found to be down regulated w.r.t XRCC-6BP1. These are reflected in rankings of 1523 (linear) and 1356 (rbf) for EXOSC3 - XRCC6BP1; EXOSC5 was found to be down regulated w.r.t XRCC-1/2/6/6BP1. These are reflected in rankings of 741 (laplace), 291 (linear) and 8 (rbf) for EXOSC5 - XRCC1; 1244 (laplace), 791 (linear) and 702 (rbf) for EXOSC5 - XRCC2; 65 (laplace), 1064 (linear) and 322 (rbf) for EXOSC5 - XRCC6; and 416 (laplace), 880 (linear) and 1434 (rbf) for EXOSC5 - XRCC6BP1. EXOSC6 was found to be down regulated w.r.t XRCC-1/2. These are reflected in rankings of 985 (linear) and 1163 (rbf) for EXOSC6 - XRCC1 and 1512 (laplace), 648 (linear) and 1458 (rbf) for EXOSC6 - XRCC2; EXOSC7 was found to be down regulated w.r.t XRCC-1/6/6BP1. These are reflected in rankings of 1510 (linear) and 1603 (rbf) for EXOSC7 - XRCC1; 584 (laplace), 1523 (linear) and 1018 (rbf) for EXOSC7 - XRCC6; and 1419 (laplace) and 876 (rbf) for EXOSC7 - XRCC6BP1. EXOSC8 was found to be down regulated w.r.t XRCC-1. These are reflected in rankings of 1373 (laplace) and 1515 (linear) for EXOSC8 - XRCC1; EXOSC9 was found to be down regulated w.r.t XRCC-1/2/6/6BP1. These are reflected in rankings of 44 (laplace), 1214 (linear) and 1410 (rbf) for EXOSC9 - XRCC1; 496 (laplace), 672 (linear) and 840 (rbf) for EXOSC9 - XRCC2; 1121 (laplace), 151 (linear) and 689 (rbf) for EXOSC9 - XRCC6 and 362 (laplace), 463 (linear) and 1741 (rbf) for EXOSC9 - XRCC6BP1.

On the right half of the table is the rankings of XRCC w.r.t EXOSC family. W.r.t EXOSC2, XRCC-2 was found to be down regulated. These are reflected in rankings of 166 (laplace), 417 (linear) and 56 (rbf) for EXOSC2 - XRCC2. W.r.t W.r.t EXOSC3, XRCC-2 was found to be down regulated. These are reflected in rankings of 166 (laplace), 417 (linear) and 56 (rbf) for EXOSC3 - XRCC2. W.r.t EXOSC5, XRCC-2 was found to be down regulated. These are reflected in rankings of 1559 (laplace) and 56 (rbf) for EXOSC5 - XRCC2. W.r.t EXOSC6, XRCC-1/2/6/6BP1 were found to be down regulated. These are reflected in rankings of 509 (laplace) and 1046(rbf) for EXOSC6 - XRCC1; 486 (laplace) and 1901(rbf) for EXOSC6 - XRCC2; 35 (linear) and 188(rbf) for EXOSC6 - XRCC6; 1295 (linear) and 366 (rbf) for EXOSC6 - XRCC6BP1. W.r.t EXOSC7, XRCC-6 was found to be down regulated. These are reflected in rankings of 1229 (linear) and 987(rbf) for EXOSC7 - XRCC1; 176 (laplace), 436 (linear) and 788 (rbf) for EXOSC7 - XRCC2; and 1074 (laplace), 242 (linear) and 288(rbf) for EXOSC7 - XRCC6. W.r.t EXOSC8, XRCC-2 was found to be down regulated. These are reflected in rankings of 13 (laplace) and 6 (rbf) for EXOSC8 - XRCC2. W.r.t EXOSC9, XRCC-2 was found to be down regulated. These are reflected in rankings of 655 (linear) and 1526 (rbf) for EXOSC9 - XRCC2 and 1206 (laplace) and 1626 (rbf) for EXOSC9 - XRCC6BP1;

Table 13 shows the derived influences which can be represented graphically, with the following influences - • XRCC w.r.t EXOSC with EXOSC-2 $->$ XRCC-2; EXOSC-3 $->$ XRCC-2; EXOSC-5 $->$ XRCC-2; EXOSC-6 $->$ XRCC-6; EXOSC-7 $->$ XRCC-1/2/6; EXOSC-8 $->$ XRCC-2; EXOSC-9 $->$ XRCC-2/6BP1; and • EXOSC w.r.t XRCC with EXOSC-2 $<-$ XRCC-1/2/6/6BP1; EXOSC-3 $<-$ XRCC-6/6BP1; EXOSC-5 $<-$ XRCC-1/2/6/6BP1; EXOSC-6 $<-$ XRCC-1/2; EXOSC-7 <- XRCC-1/6/6BP1; EXOSC-8 $<-$ XRCC-1; and EXOSC-9 $<-$ XRCC-1/2/6/6BP1.

3.1.7. RAD - FANC Cross Family Analysis

Fanconi Anemia (FA) is rare genetic disorder that happens mainly due to defects in proteins responsible for DNA repair via homologous recombination (Walden and Deans [14]). Cohn and D’Andrea [15] provides a review on the recent discoveries in the Fanconi Anemia and DNA double-strand break (DSB) repair pathways, which underscore the importance of regulated chromatin loading in the DNA damage response. Romick-Rosendale et al. [16] study the role Fanconi anemia pathway in squamous Cell Carcinoma. A review of the interplay between Fanconi anemia and homologous recombination pathways in genome integrity has been conducted by Michl et al. [17]. Liang et al. [18] observe the role of trimeric RAD51 and RAD51AP1-UAF1 complex in FANCD2. Taniguchi et al. [19] observe S-phase-specific interaction of the Fanconi anemia protein, FANCD2, with BRCA1 and RAD51. Zadorozhny et al. [20] show Fanconi anemia associated mutations destabilize RAD51 filaments and impair replication fork protection. Geng et al. [21] find RAD18-mediated ubiquitination of PCNA activates the Fanconi anemia DNA repair network. Rad18 E3 ubiquitin ligase activity mediates Fanconi anemia pathway activation and cell survival following DNA topoisomerase 1 inhibition as shown by Palle and Vaziri [22]. García-Luis and Machín [23] observe that Fanconi anaemia-like Mph1 helicase backs up RAD54 and RAD5 to circumvent replication stress-driven chromosome bridges. These findings suggest deep interactive role between the RAD and FA family. In colorectal cancer cell treated with ETC-1922159 these were found to both families were found to be down regulated. Our search engine alloted low laved numerical ranks to many of the 2^nd order combinations between the RAD - FANC family. This signifies possible synergistic mechanism between the two in CRC cells. Table 14 shows the rankings of each, with respect to the other. On the left half is the rankings of RAD family w.r.t FANC family and vice versa on the right half.

On the left half, we find, RAD-18/51/51AP1/51C/54B/54L were found to be down regulated w.r.t FANCB. These are reflected in rankings of 10 (laplace), 2219 (linear) and 625 (rbf) for RAD18 - FANCB; 247 (laplace), 73 (linear) and 610 (rbf) for RAD51 - FANCB; 479 (laplace), 1667 (linear) and 663 (rbf) for RAD51AP1 - FANCB; 769 (laplace), 536 (linear) and 887 (rbf) for RAD51C - FANCB; 468 (laplace), 133 (linear) and 438 (rbf) for RAD54B - FANCB; and 583 (laplace), 2131 (linear) and 160 (rbf) for RAD54L - FANCB. RAD-18/51/51AP1/54B/54L were found to be down regulated w.r.t FANCD2. These are reflected in rankings of 1035 (laplace), 1271 (linear) and 405 (rbf) for RAD18 - FANCD2; 885 (laplace) and 1383 (rbf) for RAD51 - FANCD2; 1734 (laplace), 644 (linear) and 1291 (rbf) for RAD51AP1 - FANCD2; 275 (laplace), 2460 (linear) and 478 (rbf) for RAD54B - FANCD2; and 493 (laplace) and 203 (rbf) for RAD54L - FANCD2; RAD-1/18/50/51/51C/54B/54L were found to be down regulated w.r.t FANCD2OS. These are reflected in rankings of 693 (laplace) and 1146 (rbf) for RAD1 - FANCD2OS; 1472 (laplace), 526 (linear) and 239 (rbf) for RAD18 - FANCD2OS; 178 (laplace) and 1534 (linear) for RAD50 - FANCD2OS; 1080 (linear) and 1226 (rbf) for RAD51 - FANCD2OS; 1297 (laplace), 977 (linear) and 1237 (rbf) for RAD51C - FANCD2OS; 475 (laplace), 1367 (linear) for RAD54B - FANCD2OS; 1227 (linear) and 252 (rbf) for RAD54L - FANCD2OS; RAD-1/18/50/51/51AP1/51C/54B/54L were found to be down regulated w.r.t FANCF. These are reflected in rankings of 1582 (linear) and 285 (rbf) for RAD1 - FANCF; 770 (laplace), 1329 (linear) and 1445 (rbf) for RAD18 - FANCF; 1403 (laplace), 1684 (linear) and 803 (rbf) for RAD50 - FANCF; 209 (laplace), 1247 (linear) for RAD51 - FANCF; 1681 (laplace), 13 (linear) for RAD51AP1 - FANCF; 1493 (laplace) and 224 (linear) for RAD51C - FANCF; 401 (laplace) and 143 (linear) for RAD54B - FANCF; for 690 (laplace), 829 (linear) for RAD54L - FANCF; RAD-1/18/50/51/51AP1/51C/54B/54L were found to be down regulated w.r.t FANCG. These are reflected in rankings of 755 (laplace), 393 (linear) and 82 (rbf) for RAD18 - FANCG; 345 (laplace), 114 (linear) and 295 (rbf) for RAD51 - FANCG; 957 (laplace), 218 (linear) and 1360 (rbf) for RAD51C - FANCG; 17 (laplace), 182 (linear) and 423 (rbf) for RAD54B - FANCG; and 1058 (laplace), 701 (linear) and 581 (rbf) for RAD54L - FANCG. RAD-18/50/51/51C/54B/54L were found to be down regulated w.r.t FANCG. These are reflected in rankings of 1693 (laplace)and 436 (rbf) for RAD18 - FANCI; 1703 (laplace) and 1458 (rbf) for RAD50 - FANCI; 1038 (laplace), 1668 (linear) and 310 (rbf) for RAD51 - FANCI; 597 (laplace) and 165 (linear) for RAD51C - FANCI; 557 (laplace) and 84 (linear) for RAD54B - FANCI; and 468 (laplace), 606 (linear) for RAD54L - FANCI.

On the right half, we find, FANCB to be down regulated w.r.t RAD-1/50/51/51AP1/51C/54B/54L. These are reflected in rankings of 1499 (laplace), 656 (linear) and 340 (rbf) for RAD1 - FANCB; 133 (laplace), 234 (linear) and 73 (rbf) for RAD50 - FANCB; 378 (linear) and 8 (rbf) for RAD51 - FANCB; 89 (laplace), 562 (linear) and 2 (rbf) for RAD51AP1 - FANCB; 460 (laplace), 187 (linear) and 86 (rbf) for RAD51C - FANCB; 486 (laplace), 891 (linear) and 568 (rbf) for RAD54B - FANCB and 41 (laplace) and 692 (rbf) for RAD54L - FANCB; FANCD2 was found to be down regulated w.r.t RAD-1/50/51/51AP1/51C/54B/54L. These are reflected in rankings of 1451 (laplace), 1605 (linear) and 796 (rbf) for RAD1 - FANCD2; 403 (linear) and 1299 (rbf) for RAD18 - FANCD2; 646 (laplace), 357 (linear) and 769 (rbf) for RAD50 - FANCD2; 591 (laplace) and 85 (rbf) for RAD51 - FANCD2; 993 (laplace) and 603 (linear) for RAD51AP1 - FANCD2; 629 (laplace), 656 (linear) and 620 (rbf) for RAD51C - FANCD2; 227 (laplace), 230 (linear) and 131 (rbf) for RAD54B - FANCD2. FANCD2OS2 was found to be down regulated w.r.t RAD-1/18/5051C/54B. These are reflected in rankings of 1455 (laplace) and 1624 (rbf) for RAD1 - FANCD2OS; 851 (laplace), 1457 (linear) and 653 (rbf) for RAD18 - FANCD2OS; 1477 (linear) and 1372 (rbf) for RAD50 - FANCD2OS; 1729 (laplace) and 779 (linear) for RAD51C - FANCD2OS; 1241 (linear) and 1637 (rbf) for RAD54B - FANCD2OS; FANCF was found to be down regulated w.r.t RAD-1/18/50/51C/54B. These are reflected in rankings of 1063 (laplace) and 196 (rbf) for RAD18 - FANCF; 1419 (linear) and 1676 (rbf) for RAD50 - FANCF; 1222 (laplace) and 1060 (linear) for RAD51 - FANCF; and 716 (linear) and 1262 (rbf) for RAD54L - FANCF; FANCG was found to be down regulated w.r.t RAD-1/50/51/51AP1/51C/54B. These are reflected in rankings of 825 (linear) and 843 (rbf) for RAD1 - FANCG; 695 (laplace), 511 (linear) and 933 (rbf) for RAD50 - FANCG; 1 (linear) and 397 (rbf) for RAD51 - FANCG; 661 (laplace), 400 (linear) and 23 (rbf) for RAD51AP1 - FANCG; 450 (laplace) and 1122 (rbf) for RAD51C - FANCG; 140 (laplace), 194 (linear) and 64 (rbf) for RAD54B - FANCG; FANCI was found to be down regulated w.r.t RAD-1/18/50/51/51AP1/51C/54B/54L. These are reflected in 897 (linear) and 664 (rbf) for RAD1 - FANCI; 1601 (laplace), 1161 (linear) and 1668 (rbf) for RAD18 - FANCI; 1133 (laplace), 1211 (linear) and 1238 (rbf) for RAD50 - FANCI; 1612 (laplace) and 1187 (rbf) for RAD51 - FANCI; 1513 (laplace), 1211 (linear) and 65 (rbf) for RAD51AP1 - FANCI; 143 (laplace), 137 (linear) and 87 (rbf) for RAD51C - FANCI; 178 (laplace), 350 (linear) and 76 (rbf) for RAD54B - FANCI; 211 (laplace) and 1128 (rbf) for RAD54L - FANCI.

Table 14 shows the derived influences which can be represented graphically, with the following influences - • RAD w.r.t FANC with RAD-18/51/51AP1/51C/54B/54L $<-$ FANCB; RAD-18/51/51AP1/54B/54L $<-$ FANCD2; RAD-1/18/50/51/51C/54B/54L $<-$ FANCD2OS; RAD-1/18/50/51/51AP1/51C/54B/54L $<-$ FANCF; RAD-1/18/50/51/51AP1/51C/54B/54L $<-$ FANCG; and RAD-18/50/51/51C/54B/54L $<-$ FANCI, and • FANC w.r.t RAD with FANCB $<-$ RAD-1/50/51/51AP1/51C/54B/54L; FANCD2 $<-$ RAD-1/50/51/51AP1/51C/54B/54L; FANCD2OS $<-$ RAD-1/18/5051C/54B; FANCF $<-$ RAD-1/18/50/51C/54B; FANCG $<-$ RAD-1/50/51/51AP1/51C/54B; FANCI $<-$ RAD-1/18/50/51/51AP1/51C/54B/54L;